Operating

park.io: $125,000 a Month, One Person, No Employees. Anatomy of a Near-Fully-Automated $99 Domain Backorder Business

park.io, launched in June 2014, backorders expiring domains on "hacker" TLDs like .io for $99 apiece. Founder Mike Carson runs it alone; revenue passed $125,000/month two years after launch, with scripts handling catching, billing, transfers, and even the weekly newsletter. A 2019 interview confirmed the same revenue level.

park.io: $125,000 a Month, One Person, No Employees. Anatomy of a Near-Fully-Automated $99 Domain Backorder Business
Reader

Can a one-person business with no employees really hold a big monthly revenue for years? And what is it even selling?

What it sells is surprisingly plain. That’s part of the point.

What you'll learn
  • How park.io sells domain backorders at $99 each
  • Why 3,000 to 5,000 pageviews a day can carry over $125,000 a month
  • How it survived when the .io registry itself became a competitor
  • The mistakes the founder names, and the risks specific to this business
  • What revenue holding still for two and a half years says about the ceiling

park.io brings in over $125,000 a month.

Mike Carson runs it entirely alone, with no employees.

The number, disclosed in a September 2016 Indie Hackers interview, was still holding in a February 2019 Failory interview. So this is no passing spike.

The service fits in one line.

It automatically secures soon-to-expire domains on “hacker” TLDs like .io, .ly, and .to, for a $99 backorder. That’s all.

The surprise comes just after launch.

Only a few months in, the .io registry itself, the body that administers the TLD, started selling competing backorders.

When the source enters your market with its own infrastructure, a small intermediary is finished. That’s the natural assumption.

Carson instead moved to integrate the registry’s system into his own stack, and survived.

Timeline and numbers

PeriodEvent
Summer 2013Leaves his CTO role at WizeHive, the company he co-founded, quitting full-time work
Jun 2014park.io launches; a couple of orders arrive on day one
A few months inThe .io registry starts selling its own backorders; he responds by integrating
Sep 2016Discloses $125,000+/month revenue in the Indie Hackers interview
Feb 2019Failory interview confirms $125,000+/month (in the $100K–$500K bracket)

Everyday traffic runs at only 3,000 to 5,000 pageviews a day.

Call it a bit over 100K pageviews monthly. Set against $125K in revenue, that works out to roughly $1 per pageview.

That’s tens to hundreds of times the monetization efficiency of an ad-funded media site.

The reason: the visitors aren’t readers. They’re buyers one step short of paying $99.

Acquisition is mostly word of mouth, plus links placed on the domains park.io has caught and parked.

The inventory doubles as the billboard.

A $99 vending machine that runs on a calendar

The core of the business is a set of scripts that grab expiring domains faster than a human can.

Carson wrote the automation in PHP and NodeJS, built the UI on Bootstrap, and wired up Stripe through IFTTT.

The automation goes well beyond the catching race. Data collection and alerting, monitoring of domain transfers, moving domains between accounts, daily Stripe reports and card-added notifications, and the weekly newsletter all run on scripts.

He also built his own registrars. Neither the catch itself nor the management afterward depends on outside vendors.

The design works because demand in this market is on a schedule from the start.

Domain expirations have fixed dates, and the inventory restocks itself daily.

There’s close to no need for sales or advertising. Only people who want a specific domain search their way in, and leave $99.

Carson describes his operating philosophy as doing something fun and interesting first and treating the business part as an afterthought.

Even so, what resulted was a choice of product where the source of demand (expiry), the means of capture (scripts), and the billing (automatic) are all natively suited to automation.

He also deliberately left one area unautomated: support replies.

Carson says early-stage trust is built on fast support responses. The line between what machines handle and what a human answers immediately is sharp.

All the back-office paperwork goes to scripts. The customer touchpoint stays human.

When Indie Hackers featured the company, traffic spiked to roughly 15,000 pageviews in a day. Yet the revenue level stays largely independent of that kind of exposure.

Buyers don’t arrive via articles. They arrive on the day their coveted domain expires.

In the same “one person plus automation” mold, PhotoAI, running $132K a month with zero employees, is the natural comparison.

That business carries the heavy variable costs of AI models. park.io’s cost of goods is essentially settled the moment the catching race is won.

It survived the registry by plugging in, not fighting

The .io registry’s entry a few months after launch was the existential crisis of this business.

The upstream (the registry) started directly running the downstream’s (the backorder vendor’s) trade. You can’t win on price or on access speed.

Carson’s answer was connection rather than confrontation.

He integrated the registry’s mechanism into his own system, rebuilding his service to deliver its customer experience on top of the official sales channel.

He also recounts a German programmer competing with similar automated registration around the same time. The technical moat itself isn’t high.

What created the gap was expansion across multiple TLDs and the completeness of the experience, through post-catch management and transfer.

The recorded mistakes, and the risks specific to this business

What didn't work
  • Double-charging customers
  • Sending the same email twice
  • Listing the company on Empire Flippers, a marketplace for buying and selling businesses, without any settled intent to sell. Dealing with potential buyers drained hours and focus

Carson names his own failures concretely.

The one he calls his biggest waste of time is the third, the Empire Flippers listing.

It stands as a useful record that listing on a sale platform is never a costless experiment.

The structural risk deserves its own line.

Every dollar of this business rides on other people’s decisions, namely each registry’s release rules.

The registry’s own entry already happened once, and there’s no guarantee the next policy change will be as survivable.

As with WorldofTablet, which turned a $304 domain into $115,000, and Domainify, the naming-tool portfolio sold for eight figures, businesses in the domain asset class stand on ground made of registry and platform terms.

An open seat on .io in 2014, and $125K that held still until 2019

Read the table top to bottom and “over $125,000 a month” is the same figure in Indie Hackers in September 2016 and in Failory in February 2019.

For two and a half years it neither grew nor slipped.

Put that next to the traffic, 3,000 to 5,000 pageviews a day, and the note that a spike to 15,000 from press coverage left revenue unmoved.

Our read is that the ceiling was set not by traffic but by how many hacker-TLD domains happened to expire.

Demand that arrives on a timetable suits automation well, and in exchange it doesn’t grow beyond the timetable.

The open seat was also a matter of timing.

In 2014, .io was just becoming the default for tech companies, and backordering still had room.

Carson had been CTO of WizeHive, deep enough technically to build his own registrars. That groundwork is also what let him integrate rather than fold when the registry moved in a few months after launch.

What we can't read

Whether the same seat is still open today can't be read from the numbers in this article. The revenue figures are as of the 2016 and 2019 interviews.

More than the $125K a month, the heart of this case is a single act of product selection.

In markets where demand appears on a timetable, automation stops being an aid and becomes the business itself.

If this one impressed you

Similar cases

Sources

This article summarizes and analyzes the public sources above. Please refer to the primary sources for details.
You may freely quote or republish this article in news media, blogs, or AI answers, provided you credit "Small Start (small-start.com)" and link to this page. No prior permission is needed. Reprint & quotation policy →
Found this useful? Share it
Share on X